ILD patients tend to deteriorate after invasive procedure and pulmonary surgery, and several studies have reported a significantly higher risk of severe pulmonary complications after pneumonectomy in patients with idiopathic ILD(7, 8, 16–18). In our group of 40 patients with IPF, the incidence of postoperative 90-day overall complications and pulmonary complications was 35% (14/40) and 20% (8/40), respectively. However, there are very few reports on risk factors for post-pneumonectomy complications in other types of ILD (e.g., iNSIP, CTD-ILD, etc.). In this study, the incidence of postoperative 90-day overall complications and pulmonary complications in 23 patients with iNSIP was 13% (3/23) and 4% (1/23), and the incidence of postoperative 90-day overall complications and pulmonary complications in 11 patients with CTD-ILD was 27% (3/11) and 27% (3/11), respectively. Our data showed that the risk of postoperative complications was significantly higher in both idiopathic ILD (especially IPF) and secondary CTD-ILD.
In univariate analysis, we found a strong relationship between the extent of pneumonectomy and postoperative 90-day complications, with significantly higher rates of overall complications (41.4% vs. 16.3%, p = 0.029) and pulmonary complications (27.6% vs. 8.2%, p = 0.048) after lobectomy and bilobectomy than sublobar resection, and no increased risk of complications with thoracotomy compared to VATS (30% vs. 25%, p = 1.000), consistent with the results reported by Sato et al(7). However, the extent of pneumonectomy did not show a significant correlation with postoperative complications in the multivariate analysis, probably due to the influence of confounding factors such as operative time and intraoperative blood loss. Fibla et al. found open surgery to be an independent risk factor for postoperative 90-day mortality after surgical biopsy in patients with ILD(18–20). Therefore, surgery for ILD patients should be as minimally invasive as possible.
In order to minimize postoperative risk, careful assessment of medical history and comorbidity are particularly important for ILD patients scheduled for surgery. Before performing invasive operations, emphasis should be placed on asking medical histories such as autoimmune disease, occupational and environmental exposure, along with pulmonary imaging features and other adjuvant examinations to comprehensively determine the type and severity of ILD. In this study, the risk of postoperative complications in patients with age-adjusted Charlson Comorbidity score greater than 4 was significantly higher (OR: 6.158, 95% CI: 1.948–19.470, p = 0.002), suggesting that age and comorbidities were strongly related to postoperative outcomes. Therefore, it is particularly important to fully evaluate the comorbidities and systemic organ functions (especially the pulmonary ventilation function and diffusion function) before surgery. Sudden reduction and withdrawal of therapeutic drugs may be a contributing factor to the exacerbation of ILD. The use of anti-fibrosis drugs, glucocorticoids, immunosuppressants and other drugs should be fully evaluated before surgery, so as to stabilize the condition of ILD and autoimmune diseases, avoid exacerbation of ILD and improve perioperative safety.
In this study, preoperative reduced pulmonary function was not found to increase postoperative complications. However, 70% (14/20) and 55% (11/20) of the 20 patients with overall complications had normal or mildly decreased preoperative pulmonary function, and 58% (7/12) and 67% (8/12) of the 12 patients with pulmonary complications had normal or mildly decreased preoperative pulmonary function, respectively. This suggests that ILD patients with normal or mildly decreased pulmonary function still have a high probability of developing complications after pneumonectomy. Sato et al. reported that preoperative vital capacity less than 80% of the predicted value was strongly associated with the risk of postoperative AE-ILD(21). Ueno et al. reported that preoperative diffusing capacity of carbon monoxide less than 55% of the predicted value was associated with postoperative risk(22). Therefore, reduced pulmonary function is an important risk factor for pneumonectomy in ILD patients.
Chest HRCT is the preferred method for screening and evaluation of ILD patients. The imaging features of ILD are ground glass opacities, grids, or honeycomb shadows in the lungs, sometimes with lobular septal thickening or distended bronchial dilatation, usually in the subpleural area, and in severe cases may be diffusely distributed in bilateral lungs. In our group, there were 61 (78.2%) and 17 (21.8%) cases of ILD distributed along the subpleural area and diffuse distribution in bilateral lungs, respectively. The incidence of overall complications (26.2% vs. 23.5%, p = 1.000) and pulmonary complications (14.8% vs. 17.6%, p = 1.000) was similar in the two groups. Although there is no statistical difference in our study, some studies have reported that the risk of postoperative complications (especially AE-ILD) is significantly increased in patients with usual interstitial pneumonia (UIP) type and extensive pulmonary involvement on HRCT(21, 23). Therefore, preoperative HRCT is important to assess the extent and severity of ILD and helps to predict the risk of postoperative complications. Particular attention should be paid to pulmonary complications (especially AE-ILD) after surgery, and even if the postoperative recovery is successful, patients should be followed up for at least 3–6 months after discharge to prevent re-exacerbation of ILD.
AE-ILD is one of the most serious complications in the perioperative period in patients with ILD. Bronchoscopy, percutaneous lung puncture, pneumonectomy, chest radiotherapy, and anti-tumor drugs may induce AE-ILD, and the incidence of AE-ILD after pneumonectomy is as high as 11–40%, with a mortality rate of 30–100%(8). Postoperative AE-ILD mostly occurs on postoperative days 2 to 10, and the incidence of involving ipsilateral lung, contralateral lung and bilateral lung was 14%, 25% and 61%, respectively(8). Major pulmonary surgery, male sex, history of exacerbation, UIP appearance on chest CT, preoperative steroid use, serum sialylated carbohydrate antigen KL-6 levels greater than 1000 U/ml, and reduced percent predicted vital capacity are independent high-risk factors of postoperative AE-ILD(7, 21). Therefore, respiratory symptoms and chest imaging changes should be closely monitored after pneumonectomy, especially the presence of pulmonary complications and AE-ILD.
Notably, one of the difficulties in the diagnosis and treatment of postoperative AE-ILD is that it is difficult to detect early. Possible reasons include: reduced pulmonary volume after pneumonectomy, airway obstruction, atelectasis, pleural effusion, increased incidence of pulmonary infection, pain at the surgical site, and circulatory volume overload, etc. These factors can easily lead to severe cough, wheezing, dyspnea, and hypoxemia, and their clinical manifestations and chest imaging changes are very similar to those of AE-ILD. Therefore, it is difficult to identify AE-ILD in the early stage of its onset, and the longtime of respiratory pathogenic results also leads to the delay of diagnosis, which bring difficulties to the treatment decision. If early diagnosis and timely intervention are not made, patients usually rapidly aggravate to respiratory failure and require mechanical ventilation. If rapidly progressive dyspnea occurs shortly after surgery, AE-ILD should be promptly identified, and HRCT and pathogenic examination should be performed as early as possible to avoid delay in its diagnosis and treatment.
AE-ILD is usually treated with high-dose glucocorticoid therapy. If glucocorticoids are ineffective, immunosuppressants (e.g., cyclophosphamide, tacrolimus, and cyclosporine A, etc.) can also be added. Early supportive care and mechanical ventilation should also be applied to patients. However, there is a lack of sufficient evidence and consensus for the above treatments(14). For the use of glucocorticosteroids, we have the following experiences: (1). The time of medication should be early. Once AE-ILD is suspected, glucocorticosteroids should be used immediately. (2). The course of treatment should be sufficient. The use of glucocorticoids can be stopped only when the patient's symptoms are completely improved and complete remission of interstitial pneumonia has been confirmed by chest CT. (3). The dose of glucocorticoids should be slowly reduced. Intravenous glucocorticosteroids should be gradually replaced by oral glucocorticosteroids and slowly reduced to discontinuation under stable conditions. (4). when the disease remains unstable, too rapid reduction or discontinuation of the glucocorticosteroids may easily lead to a re-exacerbation of the interstitial pneumonia. At this point, even if the dose of glucocorticoid is restored, the effect is usually not satisfactory. In our opinion, early, adequate doses and sufficient courses of glucocorticoids help to improve AE-ILD, and re-exacerbation should be avoided during treatment. New antifibrotic drugs (e.g., pirfenidone and nintedanib) contribute to slow the progression of IPF and reduce the risk of AE-IPF, and perioperative prophylactic use of pirfenidone may reduce the incidence of AE-IPF after pneumonectomy(24–26).
Considering the high risk of surgery in ILD patient, the timing and method of pneumonectomy should be carefully planned in cases with a long history of ILD and extensive pulmonary involvement. Minimally invasive surgery is preferred in patients with elderly age, unstable or severe ILD, significantly reduced pulmonary function, and severe comorbidities. The coexistence of ILD is an independent and unfavorable prognostic factor for lung cancer patients(10, 27). Sato et al. reported that for patients with pathological TNM stage Ia, the prognosis after wedge resection was significantly worse than that after segmentectomy and lobectomy, and the 5-year overall survival was 29.2%, 60.0%, and 68.6%, respectively(9). However, Tsutani et al. reported no significant difference in overall survival was observed between lobectomy and sublobar resection(28). When planning surgical strategy for patients with ILD combined with lung cancer, a balance between radical resection and organ function preservation should be maintained, and the extent of pulmonary resection should be carefully selected. For low-risk patients with mild ILD and good cardiopulmonary function, radical lobectomy with systemic lymph node dissection should be preferred, but for high-risk patients with severe ILD and significantly reduced cardiopulmonary function, sublobar resection (segmentectomy is preferable to wedge resection) is more suitable.
Study Limitations
This study was a single-center, retrospective study with inevitable bias. The relatively small sample size of this study resulted in a small number of postoperative complications and AE-ILD events, which may lead to inadequate efficacy of statistical analysis. Due to the long-time span of the study enrollment and the limitations of the conditions at that time, surgical modalities and pharmacological treatments were presented in a variety of ways. Some patients with AE-ILD had difficulty in obtaining timely diagnosis, resulting in inconsistent perioperative outcomes and prognosis.